Subscriber access provided by READING UNIV
Article
Is traffic still an important emitter of mono-aromatic organic compounds in European urban areas? Agnès BORBON, Anne Boynard, Therese Salameh, Alexia Baudic, Valerie Gros, Julie Gauduin, Olivier Perrussel, and Cyril Pallares Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b01408 • Publication Date (Web): 22 Nov 2017 Downloaded from http://pubs.acs.org on November 23, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 28
Environmental Science & Technology
Is traffic still an important emitter of mono-aromatic organic compounds in European urban areas?
1 2
Agnès Borbon1*, Anne Boynard2, Thérèse Salameh3ǂ, Alexia Baudic4,5 Valérie Gros4, Julie Gauduin5, Olivier Perrussel5, Cyril Pallares6
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1. Université Clermont Auvergne, CNRS, LaMP, F-63000 Clermont-Ferrand, France 2. LATMOS/IPSL, UPMC Univ. Paris 06 Sorbonne Universités, UVSQ, CNRS, Paris, France 3. Laboratoire Inter-Universitaire des Systèmes Atmosphériques (LISA), UMR CNRS 7583, Université Paris Est Créteil et Université Paris Diderot, Institut Pierre Simon Laplace, Créteil, France 4. Laboratoire des Sciences du Climat et de l’Environnement (LSCE), UMR CNRS 8212, Commissariat à l’Energie Atomique et aux énergies alternatives, Université Versailles Saint-Quentin, Institut Pierre-Simon Laplace, Gif sur Yvette, France. 5. AIRPARIF, Association Agréée de Surveillance de la Qualité de l’Air en Île-deFrance, 75004 Paris, France 6. Agence de surveillance de la qualité de l’air (ASPA), F-67300, Schiltigheim, France
18 19
ǂ
20 21
* corresponding author
[email protected] now at : IMT Lille Douai, Univ. Lille, SAGE - Département Sciences de l'Atmosphère et Génie de l'Environnement, 59000 Lille, France
1
ACS Paragon Plus Environment
Environmental Science & Technology
Page 2 of 28
22
Abstract
23
Trends of long-term observations and emission inventories suggest that traffic emissions will
24
no longer dominate the concentrations of mono-aromatic compounds (i.e. TEX - Toluene,
25
Xylenes and Ethylbenzene) in European urban areas. But the split limit between traffic and
26
other emission sector contributions such as solvent use remains tenuous. Here long-term
27
observations of an extensive set of hydrocarbons, including TEX, at traffic and urban
28
background sites in London, Paris and Strasbourg were combined to estimate the relative
29
importance of traffic emissions on TEX in every city. When analyzing the urban enhancement
30
emission ratios of TEX-to-benzene on a seasonal basis, two potential source signatures other
31
than traffic could be differentiated in all cities (1) summertime evaporation from fuel and/or
32
solvent and (2) wintertime domestic heating. However, traffic emissions still unambiguously
33
dominate the concentration levels of TEX in every city despite the reduction of their
34
emissions at exhaust pipe over the last two decades. Traffic explains between 60% and 96%
35
(at ±20%) of TEX levels while it is less clear for xylenes at some locations. Our results
36
provide a basis to evaluate regional emission inventories. The method is applicable at any
37
urban area where speciated hydrocarbon monitoring is available.
38 39
Toc Art 2
ACS Paragon Plus Environment
Page 3 of 28
40
41
Environmental Science & Technology
Key points:
•
Emission inventories and multi-year trends of long-term observations raise the
42
question of the impact of traffic emissions on TEX concentration levels in European
43
urban areas
44
•
One decade of hourly observations of hydrocarbons at traffic and urban stations in
45
Paris, London and Strasbourg is used to estimate the traffic fraction of TEX
46
concentrations
47
•
emission composition and additional wintertime domestic combustion for benzene
48
49
The seasonal signature of TEX sources is rather due to seasonal changes in traffic
•
It is demonstrated that traffic emissions still account for more than 60% of most TEX
50
concentrations regardless of the season while is less clear for xylenes at some urban
51
locations
52
3
ACS Paragon Plus Environment
Environmental Science & Technology
Page 4 of 28
53
Introduction
54
BTEX (Benzene, Toluene, Ethylbenzene and (m+p) and o-Xylenes), a sub-group of Volatile
55
Organic Compounds (VOC), still constitute a significant fraction of VOCs in urban
56
atmospheres up to 47% in volume of total Non-Methane Hydrocarbons (NMHC) after C2-
57
C9 alkanes1;
58
emission sources (transport, storage and distribution of fossil fuels, industrial and residential
59
combustion, processes) and solvent use6. For only benzene, which is known to be
60
carcinogenic7, the European air quality objective for health protection of 5 µg.m-3 (1.57 ppb
61
STP) per year is still exceeded in the center of Paris close to major roadways8. Apart from
62
their abundance and adverse health effects, BTEX have significant implications on
63
atmospheric chemistry. Together with alkenes, they show the highest ozone-forming
64
potential9,10. They have therefore been considered as key target compounds by ozone policy
65
reduction measures, especially in urban areas where ozone is usually VOC-limited11. BTEX
66
oxidation also contributes to the formation of Secondary Organic Aerosols (SOA). While high
67
uncertainties persist on SOA yields depending on smog-chamber experimental conditions,
68
BTEX show SOA yields as significant as the one of α-pinene, the most emitted and reactive
69
biogenic precursor being currently the dominant SOA source at the global scale: yield values
70
range from 1181 (benzene) to 1578 µg.m-3.ppm-1 (xylenes), respectively12,13 against
71
1826 µg.m-3.ppm-1 for α-pinene14. Recent results have suggested that BTEX could explain
72
34% of SOA production in summer15 in Paris megacity whereas such a contribution would be
73
lower when taking into account a larger set of VOC of intermediate volatility (IVOC) in Los
74
Angeles16. Like terpenes, oxidation of mono-aromatics would also lead to the rapid formation
75
of highly oxygenated molecules as recently suggested by Molteni and coworkers17.
2; 3; 4, 5
. BTEX are released into the atmosphere by either fossil fuel related
4
ACS Paragon Plus Environment
Page 5 of 28
Environmental Science & Technology
76
VOC emission reduction measures, such as emission reduction at vehicle exhaust pipe, have
77
successfully reduced their anthropogenic concentrations in cities like London18, Paris and
78
other medium-sized cities in France6, Los Angeles19 and at rural stations
79
trends of monthly average concentrations of toluene (Figure 1) and other aromatics (Figures
80
S1, S2, S3 in Supplement Information) at both urban background and traffic sites in three
81
European metropolitan areas (London, Paris and Strasbourg) support these findings.
82
Regardless of the city, all concentrations have consistently shown a clear decline at traffic
83
sites since 2000, especially in the first half of the decade. However, aromatic levels at urban
84
background sites have stayed almost constant since 2008 and concentrations at traffic sites
85
tend towards the urban background levels. This raises the question whether traffic emissions
86
still dominate TEX emissions. Up-to-date emission inventories in all three cities (see pie-
87
charts on the right hand panel of Figure 1 and Figures S1 to S3 in SI) show various features. It
88
should be noted that no local emission inventory is available for London megacity and the
89
emission distribution is adapted from the National Atmospheric Emission Inventory in UK
90
(NAEI) for the year 2000 (D. Derwent and S. Beevers’ personal communication). For toluene,
91
the road-transport sector accounts for half of its emissions in London and Paris and only 10%
92
in Strasbourg. For xylenes (Figures S1 and S2), the road-transport contribution drops down to
93
4%-33% except for o-xylene in London (57%). The solvent use and/or residential sector
94
appear as the other major source explaining between 28% and 56% of toluene emissions and
95
44 to 77% especially for xylenes. Only ethylbenzene would be still largely emitted by the
96
road-transport sector (44 – 68 %) (see Figure S3). Regional emission inventories for two
97
European cities, Paris and Zurich in 2005, also pointed to the use of solvents from all source
98
categories as the new dominant source of total VOCs at the urban scale. Finally, the
99
homogeneity of aromatic multi-year trends depicted in observations is not reflected in
100
regional emission inventories. However emission inventories for VOCs still show high
20,6
. The multi-year
5
ACS Paragon Plus Environment
Environmental Science & Technology
21
Page 6 of 28
101
uncertainties up to a factor of 4
and observations provide useful constraints for their
102
evaluation22;
103
usually point to the contribution of various sources to total measured VOC ambient levels26; 27;
104
24
105
solvent use-related activities and wood combustion for benzene26,
106
supporting source-receptor models directly imply an estimation of source/factor contributions
107
to total measured VOC concentrations rather than individual species. Towards the complexity
108
of TEX emission sources, there is still no consensus on the importance of BTEX source and
109
for traffic in particular. Moreover, the seasonality of emission changes has been poorly
110
studied for BTEX.
111
In this study, we combine long-term and continuous in-situ measurements of NMHC,
112
including BTEX, carried out in European urban areas by air quality monitoring network
113
programs over the past decade to investigate the importance of traffic emissions on mono-
114
aromatic levels at the urban scale. Due to its stringent regulation in fossil fuel and solvent
115
formulation compared to other mono-aromatic compounds, benzene will be treated here as a
116
combustion source tracer rather than a target species for the quantification of traffic
117
contribution. This study provides a scientific basis to evaluate the accuracy of regional
118
emission inventories.
119
Experimental
120
Data analysis mostly relies on continuous hourly observations of a large set of NMHC,
121
including BTEX, performed by European air quality networks in the megacities of London
122
(UK by DEFRA), Paris (France by AIRPARIF) and Strasbourg (Eastern France by ASPA).
123
The chosen period for data analysis covers the years 2008 to 2013. Special care was put on
124
data quality checks. The description of quality assurance tests is provided in Boynard et al.
23 24; 21;25
. Source-receptor models like PMF (Positive Matrix Factorization)
. BTEX usually show significant contributions in the source profiles of vehicle exhaust, 27, 24, 28
. Mathematics
6
ACS Paragon Plus Environment
Page 7 of 28
Environmental Science & Technology
125
(2014)2 and Waked et al. (2016) 6 for French data and Dollard et al. (2007)29 for UK data. The
126
purpose here is to evaluate the importance of traffic emissions relative to other anthropogenic
127
emissions. Data analysis has therefore combined observations collected at both traffic and
128
urban background sites in each city to provide perspectives from contrasted urban situations
129
non-influenced by industrial emissions. Traffic sites represent those locations under the direct
130
influence of nearby traffic emissions, usually a roadway tolerating traffic greater than 10,000
131
vehicles per day. Urban background sites are under the influence of multi-emitters, at a
132
distance of major traffic lanes of 40 m on average and are more likely representative of the
133
urban mixing of all anthropogenic emissions. Details on the database are reported in Table 1.
134
A description of the sites, meteorology and analytical devices is provided in the Supplement
135
Information (pages S5-S6). Note that the Paris data has been completed by tunnel
136
measurements performed during the PREQUALIF project 28. The number of missing data is
137
usually lower than 25% over the six-year period.
138
Results and discussion
139
Concentration levels of BTEX
140
Average concentration levels of mono-aromatics including benzene in the three metropolitan
141
areas are reported in Table S1 (2nd to 5th column) in the Supplement Information for the year
142
2011. BTEX concentration levels range between 0.03 ppb (ethylbenzene at Strasbourg Ouest
143
and 1,74 ppb (toluene at Paris-Auteuil), i.e. two orders of magnitude. Concentration levels are
144
2 to 4 times higher at the traffic sites compared to urban background sites. Concentrations of
145
benzene are quite similar between the three cities at ±0.040 ppb at traffic sites and ±0.070 ppb
146
at urban background sites. This is consistent with its stringent regulation policy at European
147
level to achieve the air quality standard of 5 µg.m-3 per year by, for instance, limiting its
148
presence at 1% vol. in fuels (Directives 98/70/EC replace by 2009/30/EC and 2003/17/EC). 7
ACS Paragon Plus Environment
Environmental Science & Technology
Page 8 of 28
149
On the opposite, the sum of BTEX load is higher in Paris by 1.4 at traffic sites to 2.4 at urban
150
background sites while concentration levels in London and Strasbourg are surprisingly quite
151
comparable. However, the atmospheric composition in BTEX is very similar between near-
152
traffic and urban background atmospheres. Together with (m+p)-xylenes, toluene (which is
153
the dominant species among BTEX) explains 70% of BTEX load. This suggests that BTEX
154
concentrations are driven by emissions with a non-visible effect of photochemical processing
155
at the background sites. For the latter we would have expected an enrichment of air mass
156
composition in longer-lived mono-aromatics like benzene given that BTEX reactivity towards
157
OH radical spans one order of magnitude30. For benzene, a domestic heating source is highly
158
suspected, especially in Strasbourg (see following sections) and as recently shown by Baudic
159
and co-workers28.
160
Seasonal source signature on TEX
161
The magnitude of TEX concentrations is affected by the proximity of emission sources, the
162
extent of photochemical processes and boundary layer conditions. To minimize those effects,
163
we use the TEX-to-benzene enhancement emission ratio (ER = ∆[TEX]/∆[benzene]) at urban
164
background sites to identify the TEX source signature on a seasonal basis in order to index
165
seasonal changes in urban enhancement ratios to potential changes in the nature and intensity
166
of their sources. The TEX-to-benzene enhancement ratio equals the slope of a linear 2-sided
167
fit of a TEX-to-benzene correlation plot. An illustration of summertime (July) and wintertime
168
(January) scatterplots is provided in Figure 2a over the six-year period at the urban
169
background sites. Urban enhancement ER of toluene-to-benzene representative of traffic,
170
domestic wood combustion and gasoline evaporation emissions are also reported to provide a
171
comparison basis, which is rather qualitative than quantitative regarding the numerous factors
172
modifying the emission composition. These three emission categories have been usually 8
ACS Paragon Plus Environment
Page 9 of 28
Environmental Science & Technology
173
identified as TEX major sources in urban areas21;
24; 28
174
values of ER is described in the Supplement Information (pages S7-S8). Emissions from
175
solvent use cannot be excluded but representative ER relative to benzene can hardly be
176
determined. The main reason is due to the diversity of solvents. Speciated emission profiles
177
usually show great variability with, for instance, toluene-to-benzene ratios spanning two
178
orders of magnitude31. Moreover, benzene, due to its toxic and carcinogen nature, has been
179
strongly limited in solvent formulations. Current standards in Europe have established limits
180
in benzene at 0.1% in cleaning products.
181
Remarkably, scatterplots show similar multiform patterns in all urban areas (Figure 2a). The
182
scatterplots lie within a domain delimited by the evaporative source ER in its upper part and
183
the domestic wood combustion ER in its lower part. Within these limits, the wintertime and
184
summertime scatterplots roughly split along the traffic ER line, suggesting the additional
185
influence of sources during both seasons despite expected seasonal changes in the
186
composition of traffic emission. In summer, the scatterplot equals (Strasbourg) or is above the
187
traffic ER line (Paris and London) and tends to the one of evaporative emissions. This would
188
suggest the presence of fugitive evaporative processes in addition to traffic. In winter, the
189
scatterplot lies between the traffic ER in its upper part and the domestic wood combustion ER
190
in its lower part. This would suggest an additional source of wood combustion for benzene27,
191
28
192
benzene mixing ratios than the ones of toluene especially in Strasbourg, a city submitted to
193
well-marked continental climatic conditions. For Strasbourg this is consistent with the
194
emission inventory (Figure 1). Note also that the multiform pattern is temperature-dependent
195
as seen in Paris and Strasbourg when color-coded with temperature (Figure S4 in Supplement
196
Information). Toluene mixing ratios stay rather constant between both seasons. One cannot
. The determination of the reported
. Note that the observed multiform patterns are rather driven by the seasonal variability of
9
ACS Paragon Plus Environment
Environmental Science & Technology
Page 10 of 28
197
exclude that the correlation slope can be affected by photochemical processing for most
198
reactive VOCs21,22 especially in summer. When applying the nighttime and daytime filters
199
values following Borbon et al. (2013)30, no significant difference is depicted suggesting that
200
emissions mainly control TEX concentrations (not shown here). It is consistent with
201
observations during MEGAPOLI-2010
202
concentrations was only visible in the Paris plume explored by the French ATR-42 aircraft
203
during the MEGAPOLI campaign in the summer 2009 at a distance above 50 km, at least,
204
from the urban center30..
205
The correlation slopes have been determined on a monthly basis at traffic and urban
206
background sites. The toluene-to-benzene ERs are reported in Figure 2b for Paris and
207
Strasbourg. As expected, a clear seasonal variability is depicted on ER time series with a
208
higher variability at urban background sites. However, both cities show an opposite pattern. In
209
Paris, ERs are similar in winter and the urban background ER significantly increases in
210
summer compared to the traffic ER. In Strasbourg, the summertime ERs are similar while the
211
urban background ER decreases compared to the traffic ER in winter. Such temporal and
212
spatial variability also suggests the potential gaining importance of summertime sources other
213
than traffic as in Paris (evaporation) and the potential gaining importance of wintertime
214
sources other than traffic as in Strasbourg (residential wood combustion).
215
Finally, seasonal scatterplots of TEX mixing ratios versus benzene mixing ratios and time
216
series of TEX-to-benzene ER suggest a seasonal change in traffic emission composition and
217
the potential influence of sources other than traffic to BTEX concentrations. The importance
218
of these additional source contributions is going to be evaluated in the next section.
219
Contribution of traffic emissions to TEX concentrations
30.
in Paris. The effect of photochemistry on TEX
10
ACS Paragon Plus Environment
Page 11 of 28
Environmental Science & Technology
220
The method consists in estimating the fraction of each TEX originating from traffic emissions
221
at the urban background sites which are representative of the urban emission mixing. This
222
fraction is calculated by dividing the TEX concentration originating from traffic emissions to
223
the measured concentration. The TEX concentration originating from traffic emissions
224
([TEX]traffic) is expressed as follows:
225
[TEX]traffic = ([tracer]traffic – [tracer]bckgd )× ERtraffic
(equation 1)
226
where [tracer]traffic is the concentration of a traffic tracer subtracted by its regional background
227
concentration ([tracer]bckgd) which could be significant for longer-lived compounds like
228
benzene28 and ERtraffic is the urban enhancement emission ratio from traffic between TEX and
229
the traffic tracer. This equation is derived from the one developed by Borbon et al. (2003)21.
230
Acetylene has been commonly used as a traffic exhaust tracer21, 32. However, a non-exhaust
231
combustion source cannot be excluded especially during the winter season regarding domestic
232
heating as for benzene
233
significant contributions in the wood burning factor solution calculated by PMF 27, 28. To take
234
into account potential additional sources to traffic and seasonal changes in their relative
235
importance, five hydrocarbon tracers have been selected in order to derive an uncertainty of
236
the TEX traffic fraction. These tracers include: four combustion products/tracers (ethylene,
237
propene, acetylene and benzene) for which a wintertime domestic heating combustion source
238
cannot be excluded and an unburned-fuel VOC (isopentane) for which a gasoline evaporation
239
source has been demonstrated in summer21, 3. In addition, source attributions for the five
240
tracers by the AIRPARIF (Paris) and NAEI (UK) emission inventories are reported in the
241
Supplement information (figure S8).
28
. In past studies, acetylene and ethylene have already shown
11
ACS Paragon Plus Environment
Environmental Science & Technology
Page 12 of 28
242
ERtraffic is determined from the slope of a two-sided linear regression fit between TEX at
243
traffic rush hours and the tracer measured at the traffic sites. For Paris, the determination of
244
ERtraffic for TEX combines ERs relative to benzene at the traffic site (Auteuil) and ER of
245
benzene relative to other tracers from the Prequalif tunnel experiment (see 6th column of
246
Table S1). While the ERtraffic from the PREQUALIF experiment was only determined during
247
the fall period (September-October 2012), it is representative of the average of expected
248
monthly ERs, the maxima and minima usually being observed in summer and winter (see
249
discussion below). While the contribution at the traffic site of additional sources can be
250
neglected, some seasonal changes in vehicle fuel composition can be expected as a result of
251
seasonal changes in fuel composition regulation and temperature-dependent evaporative
252
emissions. ERtraffic was therefore determined on a monthly basis. ERtraffic relative to the five
253
tracers values are reported in Table S1 at the three traffic sites. The monthly variability of
254
ERtraffic is significant showing minimum and maximum values in winter and summer,
255
respectively. Minimum values from near-traffic measurements found in the literature are also
256
reported for comparison. All TEX are highly correlated with the five tracers at the traffic sites
257
with an R2 higher than 0.80 and a narrow distribution along the regression slope as an
258
evidence of the traffic origin for TEX (see the example in figure S7 for toluene versus
259
benzene at traffic sites). Values of TEX-to-benzene ER are in the same range among the three
260
cities. It should be noted that ER in Paris and London are in the upper range compared to
261
Strasbourg (only toluene-to-benzene ER reported) and other ER derived from tunnel
262
measurements in Europe. This enrichment has been first pointed out by Borbon et al. (2013)
263
and would be due to two-wheeler powered vehicles (Salameh et al., to be submitted in 2017).
264
Note that the ER relative to benzene is quite similar and the greatest variability is usually
265
observed for toluene and (m+p)-xylenes. Values of ER of TEX relative to other tracers are
266
also consistent between Paris and London. 12
ACS Paragon Plus Environment
Page 13 of 28
Environmental Science & Technology
267
[tracer]bckgd is determined on an hourly basis and is derived from a linear interpolation of
268
minimum concentrations observed over a 4-day period between 2008 and 2013. The method
269
was validated by comparing the time series of interpolated background concentrations to the
270
ones observed at European rural sites in the frame of the EMEP monitoring network and
271
extracted from the EBAS database (http://ebas.nilu.no). VOC data collected twice a week is
272
available from 2008 to 2013 during a 4-hour sampling by canisters. For France, two rural sites
273
have been selected: La Tardière (LT, Western France) and Peyrusse-Vieille (PV, Southern
274
France)20. As an example, a comparison is provided on Figures S5 and S6 in the Supplement
275
Information for the five tracers for Paris and Strasbourg. Except for some periods when local
276
influences at the rural areas cannot be excluded, especially under a continental influence and
277
analytical uncertainties, the time series of interpolated urban background minimum
278
concentrations are consistent with the ones at rural background sites. Median ratios between
279
calculated backgrounds and the ones at Peyrusse-Vieille and La Tardière equal 1.25 and 0.76,
280
respectively.
281
In Figure 3, we have compared the average mixing ratios derived from the traffic
282
(ERtrafic×[tracer]) and background ([TEX]bkgd) terms to the observed mixing ratios in the case
283
of toluene at the three urban background sites. TEX background concentration was determined
284
following the 4-day minima interpolation method described previously for the five tracers. The
285
summertime average contributions of traffic, background and non-traffic terms are illustrated
286
by pie charts and the values of annual traffic contributions to measured concentrations are
287
reported below.
288
Regardless of the location, toluene concentrations show the typical seasonal variability of
289
anthropogenic VOC with maximum concentrations in winter and minimum concentrations in
290
summer. On an annual basis and regardless of the location, traffic emissions unambiguously 13
ACS Paragon Plus Environment
Environmental Science & Technology
Page 14 of 28
291
dominate toluene mixing ratios by explaining more than 70% of the observed concentrations.
292
Given the use of five tracers, the standard deviation of traffic fraction was estimated to be
293
within a ± 13-24 % range. The background contribution is significant with an annual average
294
of 7% in Strasbourg, 16% in London and 20% in Paris. Contrary to other mono-aromatics, a
295
high background level was found for toluene in Paris exceeding by almost 0.40 ppb the rural
296
ones. We suspect the background concentration being overestimated by our method. We
297
cannot exclude that the estimated background include a non-traffic source term. However,
298
when substracting the background term, traffic emissions explains more than 90% of toluene
299
concentrations. In summer, the traffic contribution remains higher than 58% (Paris, 2013) and
300
higher than 78% without accounting for the background term. This is not significantly different
301
from the annual contribution. The expected additional contribution of summertime evaporative
302
sources (like gasoline evaporation from Figure 2) is not depicted. Firstly, our apportionment
303
method (see equation 1) takes into account the seasonal variability of traffic emission
304
composition by the determination and the use of monthly ER values (Table S1). Secondly, this
305
suggests that the shape of the scatterplots reported in Figure 2a is driven rather by the seasonal
306
change of traffic emission composition and additional non-traffic sources in winter like
307
domestic combustion for the tracers. It should be noted that the sum of the traffic and
308
background terms in winter is usually higher than observed concentrations, indicating that
309
combustion sources other than traffic contribute to the tracer mixing ratios. The strongest
310
evidence is seen at the continental background site of Strasbourg. When only considering
311
isopentane as a traffic tracer in winter in Strasbourg, the traffic contribution falls by 20%
312
compared to the one by combustion tracers but remains above a 50%-contribution (not shown
313
here). However, our results suggest that the traffic source contribution has decreased by 31%
314
in Paris over the last six years. This trend is not depicted in Strasbourg and London. Further
315
analysis for the years following 2013 should be performed to confirm this trend. On an annual 14
ACS Paragon Plus Environment
Page 15 of 28
Environmental Science & Technology
316
basis, these results are consistent with emission inventories in the megacities of Paris and
317
London (Figure 1) even if the observed traffic contribution is higher. For Strasbourg, the
318
emission inventory only allocates 10% of toluene emissions to traffic (Figure 1) which would
319
be largely underestimated as suggested by our results.
320
Annual values and summertime pie charts of the traffic contribution are compiled for all other
321
mono-aromatic compounds in Table 2. The values of summertime traffic contribution are
322
usually within the standard deviation of the annual ones. While the split limit between traffic
323
and non-traffic contribution is more tenuous for >C7-aromatics (e.g. (m+p)-xylenes in
324
Strasbourg), traffic emissions would still dominate the concentrations of >C7-aromatics. The
325
summertime (and annual) contribution of traffic is usually higher than 60% and up to 96% over
326
the last six years in the three cities. One exception is the 2013-contribution with a 43%-value
327
to o-xylene concentrations in London and a 46%-contribution for (m+p)-xylenes in Strasbourg
328
in 2009. In London, the summertime traffic contribution to o-xylene concentrations has
329
decreased from 86% to 43%. The decreasing trend for (m+p)-xylenes in London is surprisingly
330
not as clear as for o-xylene. This finding is consistent with the observed change at Eltham in
331
the o-xylene-to-(m+p)-xylenes when looking at the shape of the scatterplot (not shown here)
332
Further analysis should be performed after 2013 to confirm. These results are qualitatively
333
consistent with emission inventories for ethylbenzene but usually disagree for xylenes in the
334
French cities of Paris and Strasbourg, for which the solvent use dominates by far their
335
emissions (Figure 1).
336
In Paris, the traffic contribution for the year 2010 is compared to the one extracted from a PMF
337
source-receptor study by Baudic et al.28 for the same reference year. Baudic and co-workers28
338
performed a PMF analysis on a one-year hourly VOC dataset, including the NMHC data used
339
here. They found an optimized solution of 6 factors, including a Motor Vehicle Exhaust, an 15
ACS Paragon Plus Environment
Environmental Science & Technology
Page 16 of 28
340
Evaporative Sources and a Solvent Use factor. The Evaporative Source factor is related to
341
gasoline evaporation as well as other fossil fuel storage and distribution activities. The
342
contribution of BTEX to this factor being less than 1% , the following discussion will only
343
focus on the two other factors. From the fraction of each TEX in the factor profiles and the
344
modelled concentration of each factor, the modelled contribution of the Motor Vehicle Exhaust
345
(MVE) Factor was calculated to TEX concentrations predicted by the PMF model. This
346
comparison should be interpreted with caution since (i) the MVE contribution is calculated on
347
TEX predicted concentrations (ii) PMF source profiles are assumed to be constant during the
348
whole year in the PMF calculation (iii) assumptions regarding factor identification. For the
349
latter, factor identification is consolidated by comparing to near-source measurement profiles
350
and independent source tracer such as BCff (Black Carbon from fossil fuel combustion) for the
351
MVE factor. Regarding item (i), note that observed concentrations are 30% higher than the
352
ones predicted by the PMF 6-factor solutions for toluene and higher than 20% for C8-
353
aromatics. Regarding item (ii) ER calculations have shown a clear seasonal variability
354
implying that traffic emission profiles change during the year. PMF allocates 39%±13% of
355
Motor Vehicle Exhaust and 38%±17% to Solvent Use to toluene concentrations while our
356
study allocates 79%±19% to traffic emissions on an annual basis. For C8-aromatics, PMF
357
allocates 39%±16% to Solvent Use and 36%±13% to MVE against 72%±24% to traffic
358
emissions in our study. While traffic emissions include both exhaust and gasoline evaporative
359
losses in our study, the approach developed here shows a higher contribution of traffic source
360
to aromatic emissions for all ≥C7 aromatics than the one suggested by the PMF. Given the
361
limitations of such a comparison (see above), it is expected that the factor-of-two difference of
362
traffic is the upper limit and should be lower and within the uncertainties of both approaches.
363
Except the study by Baudic (2016), there is almost no recent study reporting on VOC source
364
apportionment in European urban areas for a direct comparison. As VOC emission regulation 16
ACS Paragon Plus Environment
Page 17 of 28
Environmental Science & Technology
365
is decided at European level, we expect our results to be extrapolated to other tertiary
366
metropolitan areas in Europe which have the same anthropogenic activity pattern. However, at
367
lower latitude cities like the ones on the Mediterranean border, higher ambient temperature
368
should favor the contribution of evaporative losses.
369
Acknowledgements
370
Nadine Locoge, Stéphane Sauvage and Antoine Waked from IMT Lille Douai (France) were
371
very helpful for consolidating VOC observations from AASQA and EMEP stations.
372
Stéphane Bonneau and Alexandre Probellet at AIRPARIF and ASPA, respectively, were
373
efficient in providing the NMHC and BTEX original data from the AASQA. The authors
374
would like to thank Roy Harrison, Dick Derwent and Sean Beevers (Kings College) for the
375
information they provided on the status of VOC emission inventories for the UK and London.
376
Thérèse Salameh benefitted from an Ile de France – R2DS postdoctoral grant (2015-2016)
377
and Anne Boynard benefitted from a financial support from CNES. The authors would like to
378
thank Cathy Clerbaux for allowing the preparation of the present work.
379
Supplement Information Available. Time series of TEX concentrations in the three cities.
380
Source contributions to TEX emissions from emission inventories. Description of the sites
381
and analytical devices for VOC measurements. Determination of the toluene-to-benzene ER
382
representative of traffic, wood combustion and gasoline evaporation emissions. Time series of
383
hydrocarbon tracer concentrations at urban and rural sites for background determination.
384
BTEX average concentrations at traffic and urban background sites and ERtraffic used in source
385
apportionment calculation. Correlation plots for the determination of ERtraffic. Source
386
contributions to tracer emissions.
387
17
ACS Paragon Plus Environment
Environmental Science & Technology
Page 18 of 28
388
References
389 390 391 392
(1) Gros, V., Gaimoz, C., Herrmann, F., Custer, T., Williams, J., Bonsang, B., Sauvage, S., Locoge, N., D'Argouges, O., Sarda-Estève, R., Sciare, J., Volatile organic compounds sources in Paris in spring 2007. Part I: qualitative analysis, Environ. Chem., 2011, Vol. 8, N°1, pp 7490
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432
(2) Boynard, A., Borbon, A., Leonardis, T., Barletta, B., Meinardi, S., Blake, D.R., Locoge, N., Spatial and seasonal variability of measured anthropogenic non-methane hydrocarbons in urban atmospheres : Implication on emission ratios. Atmos. Environ., 82, 2014 pp.258-267. , (3) Salameh, T., Sauvage, S., Afif, C., Borbon A., Leonardis, T., Brioude, J., Waked, A., Locoge, N., Exploring the seasonal NMHC distribution in an urban area of the Middle East during ECOCEM campaigns: very high loadings dominated by local emissions and dynamics, Environ. Chem., 2015, 12(3), 316-328, http://dx.doi.org/10.1071/EN14154. (4) Dominutti P.A., Nogueira, T., Borbon, A., de Fatima Andrade, M., Fornaro, A., One-year of NMHCs hourly observations in São Paulo megacity: meteorological and traffic emissions effects in a large ethanol burning context, Atmos. Environ., 2016, 142, 371-382, http://dx.doi.org/10.1016/j.atmosenv.2016.08.008. (5) Wang, G., Cheng, S., Wei, W., Zhou, Y., Yao, S., Zhang, H., Characteristics and source apportionment of VOCs in the suburban area of Beijing, China, Air. Poll. Res. 2016, 7(4), 711-724, http://dx.doi.org/10.1016/j.apr.2016.03.006. (6) Waked, A., Sauvage, S., Borbon, A., Gauduin, J., Pallares, C., Vagnot, MP., Léonardis, T., Locoge, N., Multi-year levels and trends of non-methane hydrocarbon concentrations observed in ambient air in France, Atmos. Environ., 2016, 263-275, 141, https://doi.org/10.1016/j.atmosenv.2016.06.059. (7) Cocheo, V., Sacco, P., Boaretto, C., De Saeger, E., Perez Ballesta, P., Skov, H., Goelen, E., Gonzalez, N., Baeza Caracena, A., Urban benzene and population exposure, Nature Brief Communication, 2000, vol. 404. (8) AIRPARIF, Air quality annual report Surveillance et information sur la qualité de l’air en Ile de France en 2015, April 2016, 93 pages, http://www.airparif.asso.fr/_pdf/publications/bilan-2015.pdf (9) Carter W.P.L., Development of ozone reactivity scales for volatile organic compounds. Journal of Air and Waste Management Association, 1994, 44, pp. 881-899. (10) Carter W.P.L., Updated Maximum Incremental Reactivity scale for regulatory applications, Air Pollution Ressource Center and College of Engineering, Center for Environmental Research and Technology, University of California, Preliminary Report, 1998, Contract n° 95-308, Riverside, USA, 73 p.
18
ACS Paragon Plus Environment
Page 19 of 28
Environmental Science & Technology
433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448
(11) Deguillaume, L., Beekmann, M., Derognat, C., Uncertainty evaluation of ozone production and its sensitivity to emission changes over the Ile-de-France region during summer periods, J. Geophys. Res. Atmos., 2008, 113, 1–13, doi: 10.1029/2007JD009081.
449 450 451 452 453 454 455 456 457 458 459 460 461 462 463
(15) Ait-Helal, W., Borbon, A., Sauvage, S., de Gouw, J. A., Colomb, A., Gros, V., Freutel, F., Crippa, M., Afif, C., Baltensperger, U., Beekmann, M., Doussin, J.-F., Durand-Jolibois, R., I. Fronval, , N. Grand, N, Leonardis, T., Lopez, M., Michoud, V., Miet, K., Perrier, S., Prévôt, A. S. H., Schneider, J., Siour, G., Zapf, P., and Locoge, N., Volatile and intermediate volatility organic compounds in suburban Paris: variability, origin and importance for SOA formation, Atmos. Chem. Phys., 2014, 14, 10439–10464, doi:10.5194/acp-14-10439-2014
464 465 466 467 468 469 470 471 472 473 474 475 476
(18) von Schneidemesser, E., Monks, P. S. and Plass-Duelmer, C., Global comparison of VOC and CO observations in urban areas, Atmos. Environ., 2010, 44(39), 5053–5064, doi:10.1016/j.atmosenv.2010.09.010.
(12) Odum, J. R., Jungkamp, T. P. W., Griffin, R. J., Flagan, R. C. and Seinfeld, J. H.: The Atmospheric Aerosol-Forming Potential of Whole Gasoline Vapor, Science, 1997, 276(5309), 96–99, doi:10.1126/science.276.5309.96. (13) Ng, N. L., Kroll, J. H., Chan, A. W. H., Chabra, P. S., Flagan, R. C. and Seinfeld, J. H.: Secondary organic aerosol formation from m-xylene, toluene, and benzene, Atmos. Chem. Phys., 2007, 7(14), 3909–3922, doi:10.5194/acp-7-3909-2007. (14) Lee, A., Goldstein, A. H., Kroll, J. H., Ng, N. L., Varutbangkul, V., Flagan, R. C. and Seinfeld, J. H.: Gas-phase products and secondary aerosol yields from the photooxidation of 16 different terpenes, J. Geophys. Res., 2006, 111(D17), D17305, doi:10.1029/2006JD007050.
(16) Zhao, Y., Hennigan, C. J., May, A. A., Tkacik, D. S, de Gouw, J. A., Gilman, J. B., ,Kuster, W., Borbon, A., and Robinson, A.L., Intermediate-volatility organic compounds: a large source of secondary organic aerosol, Env. Sc. Tech., 2014, 48(23) Pages: 13743-13750, DOI: 10.1021/es5035188. (17) Molteni, U., Bianchi, F., Klein, F., El Haddad, I., Frege, C., Rossi, MJ., Dommen, J., Baltensperger, U., Formation of highly oxygenated organic molecules from aromatic compounds. Atmos. Chem. Phys. Discuss., 2016, doi: 10.5194/acp-2016-1126, under review.
(19) Warneke, C., de Gouw, J.A., Holloway, J.S., Peischl, J., Ryerson, T.B., Atlas, E., Blake, D., Trainer, M., Parrish, D.D., Multiyear trends in volatile organic compounds in Los Angeles, California: Five decades of decreasing emissions, J. Geophys. Res. Atmospheres, 2012, 117, D00V17, doi:10.1029/2012JD017899 (20) Sauvage, S., Plaisance, H., Locoge, N., Wroblewski, A., Coddeville, P., Galloo, J.C., Long term measurement and source apportionment of non-methane hydrocarbons in three French rural areas, Atmos. Environ., 2009, 43, 2430–2441.
19
ACS Paragon Plus Environment
Environmental Science & Technology
477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494
Page 20 of 28
(21) Borbon A., H. Fontaine, N. Locoge, M. Veillerot, J.C. Galloo, Developing receptororiented methods for non-methane hydrocarbon characterisation in urban air. Part II: source apportionment, Atmos. Environ., 2003, 37 4065–4076. (22) Warneke, C S., McKeen, A., de Gouw, J.A., Goldan, P.D., Kuster, W. C., Holloway, J.S., Williams, E.J., Lerner, B. M, Parrish, D. D., Trainer, M., Fehsenfeld, F.C., Kato, S., Atlas, E. L., Baker, A., and Blake, D. R., Determination of urban volatile organic compound emission ratios and comparison with an emissions database. J. Geophys. Res., 2007, 112, D10S47. http://dx.doi.org/10.1029/2006JD007930. (23) Coll, I., Rousseau, C., Barletta, B., Meinardi, S., and Blake, D.R., Evaluation of an urban NMHC emission inventory by measurements and impact on CTM results, Atmos. Env., 2010, 44(31), 3843-3855. (24) Gaimoz, C., Sauvage, S., Gros, V., Herrmann, F., Williams, J., Locoge, N., Perrussel, O., Bonsang, B., d'Argouges, O., Sarda-Estève, R., Sciare, J., 2011. Volatile Organic Compounds Sources in Paris in Spring 2007. Part II: Source Apportionment Using Positive Matrix Factorisation. Environ. Chem., 2011, 8, no. 1, 91–103.
495 496 497 498 499 500 501 502 503 504 505 506 507
(25) Salameh, T., Borbon, A., Afif, C., Sauvage, S., Leonardis, T., Gaimoz, C., and Locoge, N.: Composition of gaseous organic carbon during ECOCEM in Beirut, Lebanon: new observational constraints for VOC anthropogenic emission evaluation in the Middle East, Atmos. Chem. Phys., 2017, 17, 193-209, doi:10.5194/acp-17-193-2017, 2017.
508 509 510 511 512 513 514 515 516
(28) Baudic, A., Gros, V., Sauvage, S., Locoge, N., Sanchez, O., Sarda-Estève, R., Kalogridis, C., Petit, J.-E., Bonnaire, N., Baisnée, D., Favez, O., Albinet, A., Sciare, J., and Bonsang, B.: Seasonal variability and source apportionment of volatile organic compounds (VOCs) in the Paris megacity (France), Atmos. Chem. Phys., 2016, 16, 11961-11989, doi:10.5194/acp-1611961-2016
517 518 519 520
(30) Borbon, A., Gilman, J.B., Kuster, W. C., Grand, N., Chevaillier, S., Colomb, A., Dolgorouky,C., Gros, V., Lopez, M., Sarda-Esteve, R., Holloway, J., Stutz, J., Perrussel, O., Petetin, H., McKeen, S., Beekmann, M., Warneke, C., Parrish, D.D., and de Gouw, J., Emission ratios of anthropogenic VOC in northern mid-latitude megacities: observations vs.
(26) Niedojadlo, A., Becker, K. H., Kurtenbach, R., Wiesen, P., The contribution of traffic and solvent use to the total NMVOC emission in a German city derived from measurements and CMB modelling, Atmos. Environ., 2007, 41, 7108. doi:10.1016/J.ATMOSENV. (27) Lanz, V. A., Hueglin, C., Buchmann, B., Hill, M., Locher, R., Staehelin, J. and Reimann, S.: Receptor modeling of C2–C7 hydrocarbon sources at an urban background site in Zurich, Switzerland: changes between 1993–1994 and 2005–2006, Atmos. Chem. Phys., 2008, 8(9), 2313–2332, doi:10.5194/acp-8-2313-2008.
(29) Dollard, G. J., Dumitrean, P., Telling, S., Dixon, J. and Derwent, R. G., Observed trends in ambient concentrations of C2–C8 hydrocarbons in the United Kingdom over the period from 1993 to 2004, Atmos. Environ., 2007, 41, 2559–2569.
20
ACS Paragon Plus Environment
Page 21 of 28
521 522 523 524 525 526 527 528 529 530 531
Environmental Science & Technology
emission inventories in Los Angeles and Paris. J. Geophys. Res., 2013, vol. 118, 2041–2057, doi:10.1002/jgrd.50059 (31) Yuan, B., Shao, M., Lu, S., Wang, B., Source profiles of volatile organic compounds associated with solvent use in Beijing, China, Atmos. Environ., 2010, 44, 1919-1926. (32) Salameh, T., Afif, C., Sauvage, S., Borbon, A., and Locoge, N.: Speciation of NonMethane Hydrocarbons (NMHC) from anthropogenic sources in Beirut, Lebanon, Environ. Sci. Pollut. Res., 2014, 21, 10867–10877, doi:10.1007/s11356-014-2978-5.
21
ACS Paragon Plus Environment
Environmental Science & Technology
532
Page 22 of 28
Figures and Tables
533 534
535 536 537 538 539 540 541 542 543 544 545
Figure 1 : Time series of toluene monthly mean concentrations in Paris, Strasbourg and London at traffic and urban background sites since 2000 (left panel). The grey shaded area corresponds to the standard deviation of the monthly mean concentrations at the background sites. Associated apportioned emissions of toluene extracted from local emission inventories are reported on right panel (grey: road transport ; red: solvent use or residential for Strasbourg; dark blue: extraction and distribution of fossil fuel; yellow: non-industrial combustion and industrial activities for Strasbourg; light blue: other activities.
22
ACS Paragon Plus Environment
Page 23 of 28
Environmental Science & Technology
546
a)
547
548 549 550 551 552 553 554 555
b)
Figure 2: (a) Scatterplots of toluene versus benzene concentrations in January (blue) and July (red) months at London, Paris and Strasbourg urban background sites (2008-2013). Dotted lines are average toluene-to-benzene ERs representative of traffic (Aït-Helal et al., 2015), domestic wood combustion (Gaeggeller et al. 2008; Evtyugina et al., 2014; Baudic et al., 2016) and gasoline evaporation emissions (Salameh et al., 2017). (b) Time series of monthly toluene-to-benzene urban enchancement ratios (ER) derived from the slopes of correlation plots for Paris and Strasbourg at traffic and urban background sites
556
23
ACS Paragon Plus Environment
Environmental Science & Technology
Page 24 of 28
557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604
Figure 3 : Time series of toluene concentration terms in Paris, Strasbourg and London between 2008 and 2013. Data are smoothed a 30-day basis. Below time series, pie charts represent the average summertime (JJA) percent contribution of the traffic term (in grey), 24
ACS Paragon Plus Environment
Page 25 of 28
605 606 607
Environmental Science & Technology
background term (in blue) and non-traffic term (in red) to toluene concentrations. Framed numbers are the annual average contribution of traffic and its standard deviation.
25
ACS Paragon Plus Environment
Environmental Science & Technology
608 609
Page 26 of 28
Table 1: VOC data used for this study Location
Site
Type
Coord.
Network
Period
% missing data
Reference
London
Marylebone Road Eltham
Traffic Urban
51.52 N, -0.1546 W 51.45 N, 0.071 E
Hydrocarbon Automatic Network Hydrocarbon Automatic Network
2008 - 2013 2008 - 2013
23% 19%
Dollard et al., 2007 Dollard et al., 2007
Paris
Auteuil Les Halles - Crillon Guy Moquet tunnel
Traffic Urban Tunnel
48.85N 2.25E 48.86N 2.34 E 48.01N 2.65E
AIRPARIF AIRPARIF PREQUALIF experiment
2008 - 2011 24% 2008 - 2013 6% Sept-Oct. 2012 2%
Waked et al., 2016 Baudic et al., 2016
Clémenceau Strasbourg Ouest
Traffic Urban
48.35N 7.44E 48.36N 7.42E
ASPA ASPA
2008 - 2013 2008 - 2013
Boynard et al., 2014
Strasbourg
3% 16%
610 611 612 613 614
26
ACS Paragon Plus Environment
Page 27 of 28
Environmental Science & Technology
615
Aromatics
Cities
2008
2009
2010
2011
2012
2013
Paris
(m+p)-xylenes
London
Strasbourg
Paris
o-xylene
London
Strasbourg
Paris
ethylbenzene
London
Strasbourg
27
ACS Paragon Plus Environment
Environmental Science & Technology
Page 28 of 28
616
Table 2:
617 618 619 620 621 622
Table 2: Mean contribution in % of traffic emissions (in grey) to (m+p)-xylenes, o-xylene and ethylbenzene concentrations in summer (JJA) in Paris, London and Strasbourg between 2008 and 2013. Non-traffic contribution (in red) includes the background. Framed numbers represent the annual mean contribution of traffic emissions
28
ACS Paragon Plus Environment